Seepage Threatening Area Levees

Investigators studying the New Orleans flooding disaster have identified underground water seepage as a key reason certain levees collapsed during Hurricane Katrina.

A mostly invisible problem until it is too late, deep water seepage also is an urgent concern in Sacramento and throughout the Central Valley.

For years, flood control experts have had their hands full designing and maintaining above-ground levees that contain the rivers flowing across the landscape.

But now, they also are trying to get a grip on unseen waters traveling silently beneath those levees.

The Sacramento Area Flood Control Agency recently concluded a massive drilling project along levees in Natomas, involving 110 deep drilling sites, to determine the depth and extent of seepage there.

In Yolo County, flood control officials in West Sacramento are evaluating similar geotechnical data already on the books, and are planning more investigative drilling next spring.

At the same time, the state has identified underground seepage as a major concern across the Central Valley, and is working on a plan for evaluating the layers of ground and the water flows beneath levees.

In New Orleans, preliminary reviews have found that such underground flows undermined certain levees and caused them to crumble after Hurricane Katrina hit. The seepage occurred below walls of sheet pilings that had been driven 10 feet into the earthen levees, far shallower than what engineers now say is necessary to protect the city.

Les Harder, California's acting deputy director for public safety for the Department of Water Resources, was part of a National Science Foundation team that investigated the New Orleans flooding disaster. He came home shaking his head - both at the losses in the South and at the potential for losses in California.

"Do we have those cases here? You bet," Harder said of the underground seepage threat. "Have we had past failures because of it? Absolutely."

Engineers long have been aware of the possibility of seepage under levees.

Whenever water is higher on one side of a levee than the other and there is porous material in between, water will tend to move out of its regular channel, often building speed and picking up sediment because of the pressure, said Robert Bea, a professor of civil and environmental engineering at the University of California, Berkeley, who also was on the National Science team investigating in New Orleans.

"Water is really smart," Bea explained. "It wants to go from high pressure to low, and from the heavy water side to the other side."

For Sacramento and surrounding counties, awareness of the insidious seepage problem grew after the floods of 1986 and 1997.

In February 1986, a levee broke in Linda not only after the Feather River had dropped from its peak during heavy rains but also in a spot without apparent weaknesses, said Stein Buer, executive director of the Sacramento Area Flood Control Agency. Deep seepage was believed to have undermined the levee, he said.

"Here was something that came right out of the blue and shocked everyone," he recalled.

The 1986 floods prompted major levee improvement work in Sacramento, including sinking miles of slurry cut-off walls into levees on the Sacramento River in the Pocket area, as well as locations farther north.

The slurry walls were made of an impervious clay and cement mixture, and went 25 to 30 feet deep. The idea was to stop river water from seeping toward the land side through porous sandy layers within and beneath the levees.

It wasn't until the floods of 1997 that the depth of the problem really hit home.

In that event, two levees in Yuba and Sutter counties crumbled. Engineers later concluded the levees' stability likely had been undermined by deep water seepage, Harder said.

Equally troubling, engineers found evidence of underground seepage even where the 25-to 30-foot slurry walls had been installed - indicating deeper seepage than previously realized, Buer said.

That brought a round of deeper slurry cut-off walls to a new flood improvement project along the American River. Between 1999 and 2003, slurry walls reaching as deep as 80 feet were installed along almost the entire stretch of the American, from downtown to the Mayhew area near Rancho Cordova.

Buer said he and other flood officials are confident the slurry walls along the American are deep enough to stop seepage from eroding above-ground levees.

The next step, he said, is evaluating the levees in the Natomas area to assess the extent of deep seepage there. This past year, between July and October, Buer's agency spent $552,000 to have 110 holes drilled in levees in Natomas. The geotechnical samples are being studied. In February, the findings will be released and flood officials will determine where - if anywhere - deep slurry walls should go. Installing berms and relief wells are other options.

The work in Natomas comes as part of an effort to raise that area's flood protection from the current 100-year level (meaning the system is designed to handle a major storm with a 1 in 100 chance of occurring in any given year) to 200-year protection against even larger storms.

In the Pocket, 100-year protection has not been reached yet. The area has several levee erosion sites that must be repaired. In addition, three locations have been marked as having deep seepage. The U.S. Army Corps of Engineers is now deciding whether to sink slurry cut-off walls in those spots, said John Hess, chief of geotechnical and environmental engineering at the Corps of Engineers in Sacramento.

Without further studies, it is not possible to say whether much of the Pocket needs deeper slurry walls, Buer said. He noted that the American River, which can flow fast out of the mountains, tends to carry - and hence deposit - more sand and porous materials. The Sacramento River, in contrast, carries more silt and clay, which are less pervious, he said.

Deep seepage also is on flood officials' minds in West Sacramento. Ken Ruzich, manager of Reclamation District 900, which oversees West Sacramento's levees, said the district hopes to have a work plan by spring for possible exploratory drilling.

"We haven't seen any boils or visible water" to suggest seepage like New Orleans had, Ruzich said.

The state Department of Water Resources hopes to do widespread evaluations for seepage elsewhere in the Central Valley, but does not have a budget or time line for such work, Harder said.

Hess stressed that although seepage is a major concern, levees overtopping their banks remains the most common reason they fail and a giant worry for the state.